.png){kind=icon}
.png){kind=icon}
Return
On deep learning for computing the dynamic initial margin and margin value adjustment
.png){kind=icon}
PRE
AI
DOI：10.1016/j.amc.2025.129679.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A cost-effective method is provided to approximate the Dynamic Initial Margin (DIM), a crucial component in the computation of Margin Valuation Adjustment (MVA), using neural networks trained with noisy but unbiased DIM samples from single Monte Carlo paths. • The neural network is parameterized by initial market state variables and employed as a multi-output model, enabling the prediction of DIM trajectories at various monitoring times across a range of such states. • The approach demonstrates reliable performance under different interest rate models (Hull-White and CIR++) and for portfolios of varying complexity.
Keywords:
Deep learning
Initial margin
Dynamic initial margin
MVA
68T07
91G30
91G60
91G70
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.4
Papers:
2.3W
Citations:
3.3W